YOLOv5 based object detection in reel package X-ray images of semiconductor component.

The industrial manufacturing landscape is currently shifting toward the incorporation of technologies based on artificial intelligence (AI). This transition includes an evolution toward smart factory infrastructure, with a specific focus on AI-driven strategies in production and quality control. Specifically, AI-empowered computer vision has emerged as a potent tool that offers a departure from extant rule-based systems and provides enhanced operational efficiency at manufacturing sites. As the manufacturing sector embraces this new paradigm, the impetus to integrate AI-integrated manufacturing is evident. Within this framework, one salient application is AI deep learning-facilitated small-object detection, which is poised to have extensive implications for diverse industrial applications. This study describes an optimized iteration of the YOLOv5 model, which is known for its efficacious single-stage object-detection abilities underpinned by PyTorch. Our proposed "improved model" incorporates an additional layer to the model's canonical three-layer architecture, augmenting accuracy and computational expediency. Empirical evaluations using semiconductor X-ray imagery reveal the model's superior performance metrics. Given the intricate specifications of surface-mount technologies, which are characterized by a plethora of micro-scale components, our model makes a seminal contribution to real-time, in-line production assessments. Quantitative analyses show that our improved model attained a mean average precision of 0.622, surpassing YOLOv5's 0.349, and a marked accuracy enhancement of 0.865, which is a significant improvement on YOLOv5's 0.552. These findings bolster the model's robustness and potential applicability, particularly in discerning objects at reel granularities during real-time inferencing.

Metaheuristic-Based Feature Selection Methods for Diagnosing Sarcopenia with Machine Learning Algorithms.

This study explores the efficacy of metaheuristic-based feature selection in improving machine learning performance for diagnosing sarcopenia. Extraction and utilization of features significantly impacting diagnosis efficacy emerge as a critical facet when applying machine learning for sarcopenia diagnosis. Using data from the 8th Korean Longitudinal Study on Aging (KLoSA), this study examines harmony search (HS) and the genetic algorithm (GA) for feature selection. Evaluation of the resulting feature set involves a decision tree, a random forest, a support vector machine, and naïve bayes algorithms. As a result, the HS-derived feature set trained with a support vector machine yielded an accuracy of 0.785 and a weighted F1 score of 0.782, which outperformed traditional methods. These findings underscore the competitive edge of metaheuristic-based selection, demonstrating its potential in advancing sarcopenia diagnosis. This study advocates for further exploration of metaheuristic-based feature selection's pivotal role in future sarcopenia research.

Offshore MTDC Transmission Expansion for Renewable Energy Scale-up in Korean Power System: DC Highway.

In this study, we analyzed the impact of multi-terminal direct current (MTDC) system on the integration of renewable energy resources into the Korean power system. Due to the large-scale renewable energy plants planned to be integrated into the power system, line congestion is expected in the southern part of power system. Given the difficulty in constructing AC transmission lines due to social conflicts, we proposed an alternative solution using an offshore multi-terminal DC offshore transmission system. Firstly, we calculate the effective renewable energy plant generation capacity based on annual wind and solar radiation data. Next, we conduct PSS/E simulations to minimize future line congestion in the Korean power grid. The offshore terminal is designed to transfer the power generated in southern Korea and is verified using different terminal rating cases. The simulation result, including contingency analysis, demonstrate that transferring 80% of the generated renewable power achieves the best line flow condition. Therefore, the MTDC system is a possible candidate for integrating future renewable energy systems into the Korean power grid.

A Double Open-Shelled Au43 Nanocluster with Increased Catalytic Activity and Stability.

Atomically precise metal nanoclusters (NCs) are an intriguing class of crystalline solids with unique physicochemical properties derived from tunable structures and compositions. Most atomically precise NCs require closed-shells and coordinatively saturated surface metals in order to be stable. Herein, we report Au43(C≡CtBu)20 and Au42Ag1(C≡CtBu)20, which feature open electronic and geometric shells, leading to both paramagnetism (23 valence e-) and enhanced catalytic activity from a single coordinatively unsaturated surface metal. The Au-alkynyl surface motifs of these NCs form five helical stripes around the inner Au12 kernel, imparting chirality and high thermal stability. Density functional theory (DFT) calculations suggest that there are minimal energy differences between the open-shelled NCs and hypothetical closed-shell systems and that the open-shelled electronic configuration gives rise to the largest band gap, which is known to promote cluster stability. Furthermore, we highlight how coordinatively unsaturated surface metals create active sites for the catalytic oxidation of benzyl alcohol to benzaldehyde, leading to high selectivity and increased conversion. This work represents the first example of an atomically precise Au NC with a double open-shelled structure and provides a promising platform for investigating the magnetic and catalytic properties of noble metal nanoparticles.

Identification of TRPC6 as a Novel Diagnostic Biomarker of PM-Induced Chronic Obstructive Pulmonary Disease Using Machine Learning Models.

Chronic obstructive pulmonary disease (COPD) was the third most prevalent cause of mortality worldwide in 2010; it results from a progressive and fatal deterioration of lung function because of cigarette smoking and particulate matter (PM). Therefore, it is important to identify molecular biomarkers that can diagnose the COPD phenotype to plan therapeutic efficacy. To identify potential novel biomarkers of COPD, we first obtained COPD and the normal lung tissue gene expression dataset GSE151052 from the NCBI Gene Expression Omnibus (GEO). A total of 250 differentially expressed genes (DEGs) were investigated and analyzed using GEO2R, gene ontology (GO) functional annotation, and Kyoto Encyclopedia of Genes and Genomes (KEGG) identification. The GEO2R analysis revealed that TRPC6 was the sixth most highly expressed gene in patients with COPD. The GO analysis indicated that the upregulated DEGs were mainly concentrated in the plasma membrane, transcription, and DNA binding. The KEGG pathway analysis indicated that the upregulated DEGs were mainly involved in pathways related to cancer and axon guidance. TRPC6, one of the most abundant genes among the top 10 differentially expressed total RNAs (fold change ≥ 1.5) between the COPD and normal groups, was selected as a novel COPD biomarker based on the results of the GEO dataset and analysis using machine learning models. The upregulation of TRPC6 was verified in PM-stimulated RAW264.7 cells, which mimicked COPD conditions, compared to untreated RAW264.7 cells by a quantitative reverse transcription polymerase chain reaction. In conclusion, our study suggests that TRPC6 can be regarded as a potential novel biomarker for COPD pathogenesis.

A Comprehensive Study of Cyber Attack Mitigation with the Exchange of Frequency Containment Reserves Control in a Multi-Infeed Direct Current Power System.

By 2040, the Korean government aims for a penetration rate of 30-35% of the total power from renewable sources. Due to a lack of inertia, particularly in remote systems such as those on Jeju Island, these circumstances will reduce network stability. To maintain the diversity and unpredictability of RES penetration, HVDC systems with an exchange of frequency containment reserve control are utilized. An exchange of frequency containment reserves control (E-FCR) is one of the balancing arrangement concepts of HVDC systems. However, the development of E-FCR concepts is vulnerable to cyber attacks because this concept only considers one wide-area measurement for data exchange. This study established a simultaneous cyber attack operation, i.e., an attack was set at the same time as a contingency operation that affects the balancing arrangement between two regions. Multiple possibilities of cyber attack and mitigation operations were suggested according to their ability to access information in the MIDC system. Then, a cyber detection strategy was proposed through a normalized correlation concept to activate mitigation control that could enhance the frequency stability by adjusting the value of the ramp-rate deviation between two HVDC types. By simulating the Korean power system model that was implemented in PSS/E, along with a Python script, simulation results demonstrated that a cyber attack on missing data can cause severe low-frequency nadir responses, and the proposed methodology can practically detect and mitigate cyber attacks.

Identification of 2,4-Di-tert-butylphenol as a Novel Agonist for Insect Odorant Receptors.

Odorant molecules interact with odorant receptors (ORs) lining the pores on the surface of the sensilla on an insect's antennae and maxillary palps. This interaction triggers an electrical signal that is transmitted to the insect's nervous system, thereby influencing its behavior. Orco, an OR coreceptor, is crucial for olfactory transduction, as it possesses a conserved sequence across the insect lineage. In this study, we focused on 2,4-di-tert-butylphenol (DTBP), a single substance present in acetic acid bacteria culture media. We applied DTBP to oocytes expressing various Drosophila melanogaster odor receptors and performed electrophysiology experiments. After confirming the activation of DTBP on the receptor, the binding site was confirmed through point mutations. Our findings confirmed that DTBP interacts with the insect Orco subunit. The 2-heptanone, octanol, and 2-hexanol were not activated for the Orco homomeric channel, but DTBP was activated, and the EC50 value was 13.4 ± 3.0 µM. Point mutations were performed and among them, when the W146 residue changed to alanine, the Emax value was changed from 1.0 ± 0 in the wild type to 0.0 ± 0 in the mutant type, and all activity was decreased. Specifically, DTBP interacted with the W146 residue of the Orco subunit, and the activation manner was concentration-dependent and voltage-independent. This molecular-level analysis provides the basis for novel strategies to minimize pest damage. DTBP, with its specific binding to the Orco subunit, shows promise as a potential pest controller that can exclusively target insects.

Estimating vehicle speed by analyzing the acoustic frequency of dashboard camera sound.

Dashboard cameras are now widely distributed and help determine the cause of traffic accidents and vehicle speed. The simplest method to calculate the accident vehicle's speed uses a specific section's driving distance and time, determining the average speed. However, we cannot determine whether the vehicle was accelerating or decelerating. In addition, setting the reference point of the specific road section is difficult if the video is too dark or the resolution is too low. Therefore, this study aimed to calculate vehicle speed using sound data from dashboard camera videos. Herein, we suggest a method to calculate the speed of a vehicle in an accident by analyzing the engine sound frequency. Furthermore, our method was verified experimentally and in an actual case. We found that this method could obtain more information than other dashboard camera speed calculations, such as when the vehicle accelerates, decelerates, and shifts gears. However, torque converter slips increased error in the speed calculations. Therefore, we propose using the engine's sound obtained from dashboard camera videos to calculate vehicle speed if the torque converter slip is not severe.

High-Efficiency WSe2 Photovoltaic Devices with Electron-Selective Contacts.

A rapid surge in global energy consumption has led to a greater demand for renewable energy to overcome energy resource limitations and environmental problems. Recently, a number of van der Waals materials have been highlighted as efficient absorbers for very thin and highly efficient photovoltaic (PV) devices. Despite the predicted potential, achieving power conversion efficiencies (PCEs) above 5% in PV devices based on van der Waals materials has been challenging. Here, we demonstrate a vertical WSe2 PV device with a high PCE of 5.44% under one-sun AM1.5G illumination. We reveal the multifunctional nature of a tungsten oxide layer, which promotes a stronger internal electric field by overcoming limitations imposed by the Fermi-level pinning at WSe2 interfaces and acts as an electron-selective contact in combination with monolayer graphene. Together with the developed bottom contact scheme, this simple yet effective contact engineering method improves the PCE by more than five times.

Photo-Sintered Silver Thin Films by a High-Power UV-LED Module for Flexible Electronic Applications.

In recent printed electronics technology, a photo-sintering technique using intense pulsed light (IPL) source has attracted attention, instead of conventional a thermal sintering process with long time and high temperature. The key principle of the photo-sintering process is the selective heating of a thin film with large light absorption coefficients, while a transparent substrate does not heat by the IPL source. Most research on photo-sintering has used a xenon flash lamp as a light source. However, the xenon flash lamp requires instantaneous high power and is unsuitable for large area applications. In this work, we developed a new photo-sintering system using a high-power ultraviolet light emitting diode (UV-LED) module. A LED light source has many merits such as low power consumption and potential large-scale application. The silver nanoparticles ink was inkjet-printed on a polyethylene terephthalate (PET) and photo-sintered by the UV-LED module with the wavelength of 365 and 385 nm. The electrical resistivity as low as 5.44 × 10-6 Ω·cm (just about three times compared to value of bulk silver) was achieved at optimized photo-sintering conditions (wavelength of 365 nm and light intensity of 300 mW/cm2).

In Situ Ultraviolet Polymerization Using Upconversion Nanoparticles: Nanocomposite Structures Patterned by Near Infrared Light.

In this paper, we report an approach to polymerization of a nanocomposite containing UV-polymerizable organic material and inorganic, NaYbF4:Tm3+ core-based nanoparticles (NPs), which are optimized for upconversion of near infrared (NIR) to ultraviolet (UV) and blue light. Our approach is compatible with numerous existing UV-polymerizable compositions and the NaYF4: Yb, Tm3+ core-based NPs are much more stable against harsh conditions than NIR organic photo-initiators proposed earlier. The use of a core-shell design for the NPs can provide a suitable method for binding with organic constituents of the nanocomposite, while maintaining efficient NIR-to-UV/blue conversion in the NaYbF4 core. The prepared photopolymerized transparent polymer nanocomposites display upconversion photoluminescence in UV, visible and NIR ranges. We also demonstrate a successful fabrication of polymerized nanocomposite structure with millimeter/submillimeter size uniformly patterned by 980 nm irradiation of inexpensive laser diode through a photomask.

Analysis of Resistance According to Metal Ribbon Connection for Application to Interconnection of Shingled Photovoltaic Strings.

The Shingle Photovoltaic (PV) module is a new high power PV module technology manufactured by 'Dividing and ECA (Electrical Conductivity Adhesive) bonding' method for solar cell. In the case of a general PV module, a metal ribbon is soldered on the bus bar of the solar cell and connected to others. The dividing/ECA bonding technology connects the divided cells through bonding to manufacture a string. In order to make a module, the fabricated strings must be connected with Bus ribbon. The Shingled strings produced by the dividing and bonding method are not limited to the interconnection method by the metal ribbon. Also, it is not standardized for interconnections between strings. Therefore, we analyzed the characteristics of the shingled strings according to the soldering method. The characteristics of the string vary depending on the number of metal ribbons that contact the solar cell electrodes. Experimental results show that the series resistance increases significantly with fewer contacts. As a result, the efficiency of two-point contact decreased by 0.458%, four-point contact decreased by 0.048%, and eight-point contact decreased by 0.034%. This is because as the number of contacts increases, the resistance of the busbars becomes smaller and the contact resistance becomes smaller.

Design of High-Power and High-Density Photovoltaic Modules Based on a Shingled Cell String.

Building-integrated photovoltaic (BIPV) arrays, which are installed on the roofs of buildings as part of urban solar power generation, have created a demand for high-power and high-density photovoltaic (PV) modules to produce high-output power in a limited area. In this paper, a high-power PV module using a shingles technology is designed. When the vertical and horizontal dimensions of the module were 201.78 cm × 96.75 cm in the same area as that of the conventional PV module, the number of cell strips reached 390. When six 65-interconnection shingled strings were connected in series, the output power of 367.8 W was achieved. Compared with a conventional PV module of the same area, the output power was 8% greater.

Improvement in Power of Shingled Solar Cells for Photo-Voltaic Module.

This paper presents a study on the effects of heat treatment conditions on electrically conductive adhesives. Among the advantages of the shingled solar cells include larger active area and smaller current density since one of the main factors of the power loss is due to a decrease in current density. Therefore, when there is a small current, there is a benefit in regards to the power loss. The advantage of this new technique of developing photovoltaic modules is the increase of module power using the same installed area. Electrically conductive adhesives play an important role in the manufacture of shingled solar cells and understanding the effects of its curing condition is necessary to maximize its output power. Through changing the curing time and temperature, the optimized curing conditions for electrically conductive adhesives and fabricated shingled strings for development of a module could be established. Finally, we demonstrated a 500 mm × 500 mm photovoltaic module with a conventional and the other using the shingled method for purposes of comparison and a shingled module showed about 29% increase in maximum output power compared to a conventional module with the same installed area.

Optimization of Self-Cleaning Thin-Film Layer for Photovoltaic Applications.

To improve the productivity of a photovoltaic (PV) module, TiO2 thin films of different thicknesses were applied as a self-cleaning layer on soda-lime glass and a Si PV module by spray-coating a TiO2 solution. The structural, optical, and wettability characteristics of the TiO2 thin films were investigated with respect to the thickness. Thermogravimetric-differential thermal analysis, X-ray diffraction, field-emission scanning electron microscopy, contact-angle analysis, ultraviolet-visible spectroscopy, atomic force microscopy, Fourier transform infrared spectroscopy, and a solar simulator were used to analyze the prepared TiO2 thin films. The optimal thickness was determined to be 100 nm. The TiO2 thin film exhibited a self-cleaning ability even after post-annealing at 250 °C. After the self-cleaning ability was confirmed, the TiO2 thin film was applied to the PV module.

Growth of La-Doped BaSnO3 Film via Two-Step Spin Coating.

We successfully grew a uniform and high-transmittance La-doped BaSnO3 (LBSO) thin film by reducing the LBSO particle size and improving film uniformity. We report an alternating magnetic stirring and ultrasonication process that can effectively reduce the size of particles in the colloid. A two-step spin coating process can improve thin film uniformity. A two-step spin coating process consists of two main steps: aggregation and gradual evolution. Aggregation is achieved by a single coating, which causes the colloidal to particles agglomerate into uneven clusters on the substrate. The gradual evolution step improves the quality of the films by filling the pinholes left by the initial coating.

Electrically Conductive Adhesives and the Shingled Array Cell for High Density Modules.

The shingled array of solar cells has the advantages of a larger active area and smaller current density than conventional solar cells. Because the power loss is mainly driven by the decrease in current density, this new method has the benefit of increasing module power with the same installed area as used in other methods. As the electrically conductive adhesive (ECA), CA3556HF was chosen and characterized by analysis of reflectance and sheet resistance. These analyzed data show consistent and relevant results for the cell efficiency of separated and serially connected cells fabricated by means of the shingled array method. We successfully demonstrated the increase of the high density module (HDM) power by 5.1% for a 30 cm×30 cm area and the fill factor also increased by 2% compared with conventional modules.

Tribological and Electrical Properties of Chromium-Doped Carbon Films Fabricated by Unbalanced Magnetron Sputtering for Medical Stents.

Chromium-doped carbon (Cr:C) films were fabricated by using unbalanced magnetron sputtering with chromium (Cr) and graphite (C) targets. We investigated the structural, tribological, and electrical properties of the Cr:C films fabricated with various target power densities. The surface of all the Cr:C films was smooth and uniform, and the cross section showed a more compact and clear columnar structure as the target power density increased. The root mean square surface roughness increased and the contact angle on the film surface increased with the increase in target power density. Furthermore, the hardness and elastic modulus of the Cr:C films showed improvements, while the resistivity decreased with the increase in target power density. These results are associated with the ion bombardment and resputtering owing to the effects of the applied target power density.

Performance of Cu(In, Ga)Se2 Solar Cells on Zinc Sulfide Buffer Layers for Various Power Values of an Intense Pulsed Light System.

The effect of using an Intense Pulse Light system has been studied on zinc sulfide thin films and Cu(In, Ga)Se2 solar cells. The deposition of thin films on the zinc sulfide buffer layer is carried out on the glass and Cu(In, Ga)Se2 using the chemical bath deposition process. These zinc sulfide thin films were then subjected to treatment at different irradiation light intensities from 500 W to 2000 W, and then the effects on the layer were compared to a thermal annealed layer. The morphology and optical transmittance of the zinc sulfide layer were analyzed by field emission scanning electron microscopy and ultraviolet-visible spectrophotometry, respectively. This methodology was also applied to fabricate and investigate the efficiency, short-circuit current density, and external quantum efficiencies of the solar cells. This analysis shows that the treatments significantly change the properties of the zinc sulfide buffer layer and performance of the Cu(In, Ga)Se2 thin film solar cells.

Characteristics of Vanadium Oxide Thin Films Fabricated by Unbalanced Magnetron Sputtering for Smart Window Application.

Vanadium oxide (VOx) thin films were deposited by an unbalanced magnetron (UBM) sputtering system with a vanadium metal target and O2 reaction gas, and thermally treated at various annealing temperatures. In this work, the structural, electrical, and optical properties of the fabricated VOx films with various annealing temperatures were experimentally investigated. The UBM sputter grown VOx thin films exhibited amorphous structure, and had a very weak peak of V2O5 (002) owing to very thin films. However, the crystallite size of VOx films increased with increasing annealing temperature. The surface roughness of VOx films and average transmittance decreased with increasing annealing temperature. The resistivity of VOx films also decreased with increasing annealing temperature, while the electrical properties of films improved.